# DataSentry

> DataSentry is an open-source MCP server for analytics and data. Local-first data-quality scans with evidence and previewable, reversible repairs. It runs locally on your machine and is started by your AI client.

- **Kind:** MCP server
- **Pricing:** Open source
- **Platforms:** CLI, Web, self-hosted, Linux, macOS, Windows
- **Website:** https://jackxiaozhiren.github.io/datasentry/
- **Repository:** https://github.com/jackxiaozhiren/datasentry
- **MCP Registry name:** io.github.Jackxiaozhiren/datasentry
- **Categories:** analytics and data, databases
- **Listing updated:** 2026-11-15

## About

Most data-quality tools need you to know the rules first. DataSentry starts earlier: it scans a file with 39 deterministic detectors, attaches samples, affected ratios and confidence to each issue, and produces a six-dimension quality score covering completeness, validity, uniqueness, consistency, integrity and timeliness.

Repairs follow a conservative loop of propose, preview, apply to a copy, verify and roll back, and the original file is never overwritten. Detection and scoring do not depend on a language model; an LLM, optionally a local Ollama one, may propose a repair but people approve state-changing actions. The core runs locally on DuckDB.

It reads CSV, Parquet, JSONL, XLSX, DuckDB, SQLite, PostgreSQL, MySQL and cloud objects, and writes JSON, Markdown, HTML, JUnit and SARIF reports. The MCP server runs over stdio with the command datasentry mcp.

## Features

- **39 detectors:** Finds missing values, invalid formats, duplicates, outliers and referential problems.
- **Evidence per issue:** Each finding carries samples, ratios and confidence.
- **Reversible repair:** Previews changes and applies them to a copy, then re-scans to verify.
- **Drift history:** Tracks schema, row-count, score and issue-distribution drift.
- **CI-friendly outputs:** JUnit and SARIF reports and quality gates for pipelines.
- **Optional AI:** LLM help is optional and can stay on the machine with Ollama.

## Good for

- Scan an unfamiliar CSV for problems before loading it
- Block severe data issues in a CI job
- Repair a file without touching the original
- Compare data quality between two runs

## Install

### Claude Code (Terminal)

```sh
claude mcp add datasentry -- uvx datasentry-ai
```

### Claude Desktop (claude_desktop_config.json)

```json
{
  "mcpServers": {
    "datasentry": {
      "command": "uvx",
      "args": [
        "datasentry-ai"
      ]
    }
  }
}
```

### Cursor (~/.cursor/mcp.json)

```json
{
  "mcpServers": {
    "datasentry": {
      "command": "uvx",
      "args": [
        "datasentry-ai"
      ]
    }
  }
}
```

### VS Code (.vscode/mcp.json)

```json
{
  "servers": {
    "datasentry": {
      "type": "stdio",
      "command": "uvx",
      "args": [
        "datasentry-ai"
      ]
    }
  }
}
```

### Windsurf (~/.codeium/windsurf/mcp_config.json)

```json
{
  "mcpServers": {
    "datasentry": {
      "command": "uvx",
      "args": [
        "datasentry-ai"
      ]
    }
  }
}
```

### Other clients (your client's MCP settings (most use this shape))

```json
{
  "mcpServers": {
    "datasentry": {
      "command": "uvx",
      "args": [
        "datasentry-ai"
      ]
    }
  }
}
```

## FAQ

### Does it need an LLM?

No. Detection and scoring are deterministic and the LLM is optional.

### Does my data leave my machine?

The core scan runs locally with DuckDB, and a local Ollama provider can keep optional AI steps local too.

### How do I try it?

Install datasentry-ai from PyPI and run datasentry demo for a synthetic walkthrough.

### Can it overwrite my file?

No. Repairs are applied to a copy and can be rolled back.

### Is DataSentry free?

Yes. DataSentry is open source, so you can use and self-host it for free.

### How do I add DataSentry to Claude Code?

Run this command in your terminal: claude mcp add datasentry -- uvx datasentry-ai

### What are the best alternatives to DataSentry?

The closest alternatives to DataSentry are Agent-Native Analytics, Google Ads + Meta Ads + SEO/GEO MCP, Tapetide - Indian Stock Market MCP, Midplane and World Monitor.

## Alternatives

- [Agent-Native Analytics](https://ai-powered.com/mcp/agent-native-analytics): Agent-Native Analytics is an open-source MCP server for analytics and data.
- [Google Ads + Meta Ads + SEO/GEO MCP](https://ai-powered.com/mcp/google-ads-meta-ads-seo-geo-mcp): Google Ads + Meta Ads + SEO/GEO MCP is a freemium MCP server for marketing and seo.
- [Tapetide - Indian Stock Market MCP](https://ai-powered.com/mcp/tapetide-indian-stock-market-mcp): Tapetide - Indian Stock Market MCP is a freemium MCP server for finance and payments.
- [Midplane](https://ai-powered.com/mcp/midplane): Midplane is an open-source MCP server for databases.
- [World Monitor](https://ai-powered.com/mcp/world-monitor): World Monitor is a freemium MCP server for analytics and data.
- [CompanyData](https://ai-powered.com/mcp/companydata): CompanyData is a paid MCP server for analytics and data.
- [ChartAnt](https://ai-powered.com/mcp/chartant): ChartAnt is a freemium MCP server for design and media.
- [Quantivex Research](https://ai-powered.com/mcp/quantivex-research): Quantivex Research is a free MCP server for finance and payments.

---
Source: https://ai-powered.com/mcp/datasentry
